Out-patient hysteroscopy in asymptomatic postmenopausal women.

OBJECTIVE: To determine the prevalence of uterine and endometrial abnormalities in normal postmenopausal women and assess the accuracy of subjective hysteroscopic appearances and endometrial histology following 12 weeks of hormonal treatment.

DESIGN: A multicentered randomized double-blind placebo-controlled trial in which the volunteers received conjugated equine estrogen 0.625 mg, selective estrogen receptor modulator (in one of two doses) or placebo.

SETTING: Out-patient endoscopy unit in a large teaching hospital.

METHODS: Out-patient hysteroscopy and endometrial biopsy.

RESULTS: Forty-eight women had a hysteroscopy and endometrial biopsy: eight (16.6%; 95% confidence interval, CI 6-27%) women had intrauterine polyps and 38 (79.2%) women had normal hysteroscopies. Thirty-five women had a repeat endometrial assessment with hysteroscopy and, for the detection of proliferative endometrium (prevalence 22.8%), a sensitivity of 87.5%, a specificity of 74%, a negative predictive value of 95% and a positive predictive value of 50% were observed. There was a good proportion of agreement, 0.77 (95% CI 0.63-0.91), but a kappa score of 0.486 revealed only a moderate level of agreement. The likelihood ratios for proliferative endometrium were: LHR+ = 3.38 (fair), and LHR- = 0.17 (moderate).

CONCLUSIONS: Hysteroscopic assessment of the uterine cavity is efficient in the detection of pathological intrauterine lesions, but is only moderately successful in determining physiological changes in the endometrium. This study defines a standard of observational statistics for out-patient hysteroscopy in relation to normal endometrial histology in postmenopausal women who may have been ingesting exogenous estrogens.